Factor and figure management

Please click here for a .html version of the assignment

(It may be slow in loading due to the size of the document)

Loading libraries

Let us load the necessary library required for this homework.

suppressPackageStartupMessages(library(tidyverse))
## Warning: replacing previous import by 'tibble::as_tibble' when loading
## 'broom'
## Warning: replacing previous import by 'tibble::tibble' when loading 'broom'
suppressPackageStartupMessages(library(gapminder))
suppressPackageStartupMessages(library(gridExtra))

Part 1: Factor management

We shall be using the gapminder data set for this exercise.

Drop Oceania:

We begin by removing the rows for the continent Oceania from the gapminder data set. Before we do this, let us use the str() function to check the number of factors in the gapminder data set and the number of level each factor has.

gapminder %>% #loads the gapminder data set and pipes it into the function in the next line
  str() 
## Classes 'tbl_df', 'tbl' and 'data.frame':    1704 obs. of  6 variables:
##  $ country  : Factor w/ 142 levels "Afghanistan",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ continent: Factor w/ 5 levels "Africa","Americas",..: 3 3 3 3 3 3 3 3 3 3 ...
##  $ year     : int  1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 ...
##  $ lifeExp  : num  28.8 30.3 32 34 36.1 ...
##  $ pop      : int  8425333 9240934 10267083 11537966 13079460 14880372 12881816 13867957 16317921 22227415 ...
##  $ gdpPercap: num  779 821 853 836 740 ...

This ouput shows that the gapminder data set has two factors; country and continent. Country has 142 levels, while continent has 5 levels. Also we see that this data set has 1704 rows and 6 columns. We can visualized this data using a barchart as shown below;

gapminder %>%  # loads the gapminder data and pipes it into the next line
  ggplot(aes(continent)) + geom_bar() + # use ggplot to produce a barhcart
  ggtitle("The number of observations for each continent")

Let us arrange this data according to the population in ascending and plot a barchart of the new data set. We want to see if the arrange() function has an effect on the plot.

ArrGap <- gapminder %>%  # loads the gapminder data and pipes it into the next line
            arrange(pop) # arranges the data according to the population in ascending order
 
# display few rows of data
ArrGap %>%  # loads the ArrGap data and pipes it into the next line
  head() %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap
Sao Tome and Principe Africa 1952 46.471 60011 879.5836
Sao Tome and Principe Africa 1957 48.945 61325 860.7369
Djibouti Africa 1952 34.812 63149 2669.5295
Sao Tome and Principe Africa 1962 51.893 65345 1071.5511
Sao Tome and Principe Africa 1967 54.425 70787 1384.8406
Djibouti Africa 1957 37.328 71851 2864.9691 
# plotting the data
 ArrGap %>% 
  ggplot(aes(continent)) + geom_bar() + # use ggplot to produce a barhcart 
  ggtitle("The number of observations for each continent")

Observe from the table and figures above that the arrange() function re-arranges the rows of the data but this re-arrangement does not affect the figure produced after arranging the data.

Now, let us extract out data from the gapminder data set.

# create a subset of gapminder data that does not contain Oceania, call this data set 'GapNoOcean'
GapNoOcean <- gapminder %>% #loads the data set and pipes it into the function in the next line
        filter(continent != "Oceania")


# displaying table nicely
head(GapNoOcean,15) %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap
Afghanistan Asia 1952 28.801 8425333 779.4453
Afghanistan Asia 1957 30.332 9240934 820.8530
Afghanistan Asia 1962 31.997 10267083 853.1007
Afghanistan Asia 1967 34.020 11537966 836.1971
Afghanistan Asia 1972 36.088 13079460 739.9811
Afghanistan Asia 1977 38.438 14880372 786.1134
Afghanistan Asia 1982 39.854 12881816 978.0114
Afghanistan Asia 1987 40.822 13867957 852.3959
Afghanistan Asia 1992 41.674 16317921 649.3414
Afghanistan Asia 1997 41.763 22227415 635.3414
Afghanistan Asia 2002 42.129 25268405 726.7341
Afghanistan Asia 2007 43.828 31889923 974.5803
Albania Europe 1952 55.230 1282697 1601.0561
Albania Europe 1957 59.280 1476505 1942.2842
Albania Europe 1962 64.820 1728137 2312.8890

Let us use the str() function to check the factors and their levels of our new data set.

GapNoOcean %>% #loads the data set and pipes it into the function in the next line
  str() 
## Classes 'tbl_df', 'tbl' and 'data.frame':    1680 obs. of  6 variables:
##  $ country  : Factor w/ 142 levels "Afghanistan",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ continent: Factor w/ 5 levels "Africa","Americas",..: 3 3 3 3 3 3 3 3 3 3 ...
##  $ year     : int  1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 ...
##  $ lifeExp  : num  28.8 30.3 32 34 36.1 ...
##  $ pop      : int  8425333 9240934 10267083 11537966 13079460 14880372 12881816 13867957 16317921 22227415 ...
##  $ gdpPercap: num  779 821 853 836 740 ...

After removing the rows for the countries in Oceania from the gapminder data set, the new data set has 1680 rows, although, the number of factors in the new data frame still remains 5. Let us check the levels in the continent factor again using the level() function.

levels(GapNoOcean$continent)  # displays the number of levelsof the continent factor
## [1] "Africa"   "Americas" "Asia"     "Europe"   "Oceania"

This also shows that there are 5 level, it shows both the used and unused levels.

GapNoOcean %>% #loads the data set and pipes it into the function in the next line
  ggplot(aes(continent)) + geom_bar() +  # plots a barchart of the data
  scale_x_discrete(drop=FALSE) + ggtitle("The number of observations for each continent")

This plot shows that the continent Oceania is still a level in the data, althogh it has no entries. Let us arrange this new data set according to the number of observations in ascending order.

GapNoOcean %>%  #loads the data set and pipes it into the function in the next line
  arrange(pop) %>%  # arranges the data according to the population size in ascending order
  ggplot(aes(continent)) + geom_bar()  + # plots barchart
  ggtitle("The number of observations for each continent")

After arranging re-arranging the data, the Oceania continent is dropped from the barchart.

Let us remove unused levels in the continent.

GapNoOceanD <- GapNoOcean %>%
                droplevels() # drops the unused levels

str(GapNoOceanD) # displays the structure of the data
## Classes 'tbl_df', 'tbl' and 'data.frame':    1680 obs. of  6 variables:
##  $ country  : Factor w/ 140 levels "Afghanistan",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ continent: Factor w/ 4 levels "Africa","Americas",..: 3 3 3 3 3 3 3 3 3 3 ...
##  $ year     : int  1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 ...
##  $ lifeExp  : num  28.8 30.3 32 34 36.1 ...
##  $ pop      : int  8425333 9240934 10267083 11537966 13079460 14880372 12881816 13867957 16317921 22227415 ...
##  $ gdpPercap: num  779 821 853 836 740 ...
levels(GapNoOceanD$continent) # shows the levels in the continent factor
## [1] "Africa"   "Americas" "Asia"     "Europe"

It now shows that the continent factor has only 4 levels.

There are other ways we can drop unused levels, let us check them out!

forcats::fct_drop(GapNoOcean$continent) %>% # drops unused levels
  levels() # shows the levels in the continent factor
## [1] "Africa"   "Americas" "Asia"     "Europe"
fct_drop(GapNoOcean$continent) %>%  # drops unused levels
  levels()  # shows the levels in the continent factor
## [1] "Africa"   "Americas" "Asia"     "Europe"

Reorder the levels of the country or continent

Let us start by plotting a barchart of the GapNoOceanD data.

GapNoOceanD %>%  # loads the data set and pipes it into the next line
  ggplot(aes(continent)) + geom_bar() +  # plots the barchart
  scale_x_discrete(drop=FALSE)  + ggtitle("The number of observations for each continent")

The continent factor can be reordered in descending order of freqency on the barchart;

GapNoOceanD$continent %>% # loads the data set and pipes it into the next line
  fct_infreq() %>% # reorders level in descending order of frequency
   qplot()  + ggtitle("The number of observations for each continent in descending order") # plots barchart

This ordering is reversed using the forcats::fct_rev() function.

GapNoOceanD$continent %>% # loads the data set and pipes it into the next line
  forcats::fct_infreq() %>%  # reorders level in descending order of frequency
  forcats::fct_rev() %>%  # reverses the previous order
   qplot()  + ggtitle("The number of observations for each continent after reversing level")

Let us present a boxplot of the reordering of continents based on some principled summary of pop.

GapNoOceanD %>%
  ggplot( aes(continent,pop)) +   geom_boxplot() +
  scale_y_log10()  + ggtitle("Boxplot of population for each continent ordered by the median population")

GapNoOceanD %>%  
  group_by(continent) %>%
  mutate(median = median((pop))) %>%
  arrange(median) %>%
  ggplot( aes(continent,pop)) +   geom_boxplot(fill='pink') +
  scale_y_log10()  + ggtitle("Boxplot of population for each continent ordered by the median population")

GapNoOceanD %>%
  mutate(continent= fct_reorder(continent,pop)) %>%
  ggplot( aes(continent,pop)) +   geom_boxplot(fill='pink') +
  scale_y_log10()  + ggtitle("Boxplot of population for each continent ordered by the median population")

  • Reorder by median population of each continent
GapNoOceanD %>% # loads data and pipes it to the next line
  mutate(continent= fct_reorder(continent,pop)) %>%   # reorders the data
  ggplot( aes(continent,pop)) +   geom_boxplot() + # plots a boxplot
  scale_y_log10()  + ggtitle("Population of each continent reordered by median population")

  • Reorder by mean population of each continent
GapNoOceanD %>% # loads data and pipes it to the next line
  mutate(continent= fct_reorder(continent,pop, mean)) %>%  # reorders the data
  ggplot( aes(continent, pop)) +   geom_boxplot() +  # plots a boxplot
  scale_y_log10()  + ggtitle("Population of each continent reordered by mean population")

  • Reorder by maximum population of each continent
GapNoOceanD %>% # loads data and pipes it to the next line
  mutate(continent= fct_reorder(continent,pop, max)) %>%  # reorders the data
  ggplot( aes(continent, pop)) +   geom_boxplot() +  # plots a boxplot
  scale_y_log10()  + ggtitle("Population of each continent reordered by maximum population")

  • Reorder by minimum population of each continent
GapNoOceanD %>% # loads data and pipes it to the next line
  mutate(continent= fct_reorder(continent,pop, min )) %>%   # reorders the data
  ggplot( aes(continent,pop)) +   geom_boxplot() +  # plots a boxplot
  scale_y_log10() + ggtitle("Population of each continent reordered by minimum population")

Part 2: File I/O

Reading and writing data set into a file

First, let us extract the data set to be used for this exercise from the gapminder data set.

NewGap <- gapminder %>%  # loads data and pipes it to the next line
            filter(continent == "Europe"|continent == "Asia"|continent == "Africa"|continent == "Oceania") %>%
            arrange(gdpPercap) 

str(NewGap) # displays structure of new data set
## Classes 'tbl_df', 'tbl' and 'data.frame':    1404 obs. of  6 variables:
##  $ country  : Factor w/ 142 levels "Afghanistan",..: 28 28 74 53 28 42 88 74 18 42 ...
##  $ continent: Factor w/ 5 levels "Africa","Americas",..: 1 1 1 1 1 1 3 1 1 1 ...
##  $ year     : int  2002 2007 1952 1952 1997 1952 1952 1957 1952 1957 ...
##  $ lifeExp  : num  45 46.5 42.1 32.5 42.6 ...
##  $ pop      : int  55379852 64606759 748747 580653 47798986 1438760 20092996 813338 2445618 1542611 ...
##  $ gdpPercap: num  241 278 299 300 312 ...
# display few rows of new data set
NewGap %>% # loads data and pipes it to the next line
    head() %>%
    knitr::kable()
country continent year lifeExp pop gdpPercap
Congo, Dem. Rep. Africa 2002 44.966 55379852 241.1659
Congo, Dem. Rep. Africa 2007 46.462 64606759 277.5519
Lesotho Africa 1952 42.138 748747 298.8462
Guinea-Bissau Africa 1952 32.500 580653 299.8503
Congo, Dem. Rep. Africa 1997 42.587 47798986 312.1884
Eritrea Africa 1952 35.928 1438760 328.9406

Let us create a boxplot of the extracted data and name it Default.

Default <-NewGap %>% # loads data and pipes it to the next line
            ggplot( aes(continent,gdpPercap)) +   geom_boxplot(fill='green') + # plots a boxplot
            scale_y_log10()  + ggtitle("Without factor re-order") # scales the y-axis and put title on the graph

Default

Now, we reorder the factor levels and create another boxplot.

NewGapReorder <-NewGap %>%  # loads data and pipes it to the next line
  mutate(continent= fct_reorder(continent,gdpPercap,max))   # reorders the level of the continent using the max population

P <- NewGapReorder %>%  # loads data and pipes it to the next line
  ggplot( aes(continent,gdpPercap)) +   geom_boxplot(fill='pink') + # produces a boxplot
  scale_y_log10()  + ggtitle("Re-ordering factor by max. GdpPercap")

# put the two boxplots side-by-side
grid.arrange(Default,P,ncol=2,top="Comparing the Boxplot of GdpPerCap for the continents with and without factor re-order")

Next, we display the ordering of the levels of continent.

levels(NewGapReorder$continent)  # displays the levels of the selected factor the way the are ordered
## [1] "Africa"   "Oceania"  "Europe"   "Asia"     "Americas"

Let us create another plot of the reorder data and assign it to the variable PP. This plot will be places side-by-side to the new plot generated from the data read from files in order to determine which of the writing and reading format preserves ordering.

GapOrdered <-NewGapReorder %>%  # loads data and pipes it to the next line
  mutate(continent= fct_reorder(continent,gdpPercap,max)) %>%  # reorder continent by max population
  ggplot( aes(continent,gdpPercap)) +   geom_boxplot(fill='pink') +  # produces boxplot
  scale_y_log10()  + ggtitle("Data written to file")

read_csv() and write_csv()

Writing the reordered data to file.

write_csv(NewGapReorder,"NewGap") # writes the gapminder data set into a file called 'MyGapMinder', in .csv format

Reading back from file.

# reading data from file
GapCSV <- read_csv("NewGap")  # reads the gapminder file in csv format
## Parsed with column specification:
## cols(
##   country = col_character(),
##   continent = col_character(),
##   year = col_integer(),
##   lifeExp = col_double(),
##   pop = col_integer(),
##   gdpPercap = col_double()
## )
# displays the structure
GapCSV %>%
  str()
## Classes 'tbl_df', 'tbl' and 'data.frame':    1404 obs. of  6 variables:
##  $ country  : chr  "Congo, Dem. Rep." "Congo, Dem. Rep." "Lesotho" "Guinea-Bissau" ...
##  $ continent: chr  "Africa" "Africa" "Africa" "Africa" ...
##  $ year     : int  2002 2007 1952 1952 1997 1952 1952 1957 1952 1957 ...
##  $ lifeExp  : num  45 46.5 42.1 32.5 42.6 ...
##  $ pop      : int  55379852 64606759 748747 580653 47798986 1438760 20092996 813338 2445618 1542611 ...
##  $ gdpPercap: num  241 278 299 300 312 ...
##  - attr(*, "spec")=List of 2
##   ..$ cols   :List of 6
##   .. ..$ country  : list()
##   .. .. ..- attr(*, "class")= chr  "collector_character" "collector"
##   .. ..$ continent: list()
##   .. .. ..- attr(*, "class")= chr  "collector_character" "collector"
##   .. ..$ year     : list()
##   .. .. ..- attr(*, "class")= chr  "collector_integer" "collector"
##   .. ..$ lifeExp  : list()
##   .. .. ..- attr(*, "class")= chr  "collector_double" "collector"
##   .. ..$ pop      : list()
##   .. .. ..- attr(*, "class")= chr  "collector_integer" "collector"
##   .. ..$ gdpPercap: list()
##   .. .. ..- attr(*, "class")= chr  "collector_double" "collector"
##   ..$ default: list()
##   .. ..- attr(*, "class")= chr  "collector_guess" "collector"
##   ..- attr(*, "class")= chr "col_spec"
# display in table
GapCSV %>%
  head() %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap
Congo, Dem. Rep. Africa 2002 44.966 55379852 241.1659
Congo, Dem. Rep. Africa 2007 46.462 64606759 277.5519
Lesotho Africa 1952 42.138 748747 298.8462
Guinea-Bissau Africa 1952 32.500 580653 299.8503
Congo, Dem. Rep. Africa 1997 42.587 47798986 312.1884
Eritrea Africa 1952 35.928 1438760 328.9406

Observe that the structure is not the same as the data written to file, the factors continent and country are now character, although the table is still the same.

P2 <- GapCSV %>%
  ggplot( aes(continent,gdpPercap)) +  geom_boxplot() +
  scale_y_log10() + ggtitle("Data read from file")

grid.arrange(GapOrdered,P2,ncol=2,top="Comparing the boxplots otained from data written to file and the one from data read from file")

read_delim() and write_delim()

Writing to file:

write_delim(NewGapReorder,"NewGap.txt",delim = "$")  # writes the data to file

Reading from file:

GapTxt <- read_delim("NewGap.txt", delim="$")  # reads data from file
## Parsed with column specification:
## cols(
##   country = col_character(),
##   continent = col_character(),
##   year = col_integer(),
##   lifeExp = col_double(),
##   pop = col_integer(),
##   gdpPercap = col_double()
## )
str(GapTxt) # display the structure of the data
## Classes 'tbl_df', 'tbl' and 'data.frame':    1404 obs. of  6 variables:
##  $ country  : chr  "Congo, Dem. Rep." "Congo, Dem. Rep." "Lesotho" "Guinea-Bissau" ...
##  $ continent: chr  "Africa" "Africa" "Africa" "Africa" ...
##  $ year     : int  2002 2007 1952 1952 1997 1952 1952 1957 1952 1957 ...
##  $ lifeExp  : num  45 46.5 42.1 32.5 42.6 ...
##  $ pop      : int  55379852 64606759 748747 580653 47798986 1438760 20092996 813338 2445618 1542611 ...
##  $ gdpPercap: num  241 278 299 300 312 ...
##  - attr(*, "spec")=List of 2
##   ..$ cols   :List of 6
##   .. ..$ country  : list()
##   .. .. ..- attr(*, "class")= chr  "collector_character" "collector"
##   .. ..$ continent: list()
##   .. .. ..- attr(*, "class")= chr  "collector_character" "collector"
##   .. ..$ year     : list()
##   .. .. ..- attr(*, "class")= chr  "collector_integer" "collector"
##   .. ..$ lifeExp  : list()
##   .. .. ..- attr(*, "class")= chr  "collector_double" "collector"
##   .. ..$ pop      : list()
##   .. .. ..- attr(*, "class")= chr  "collector_integer" "collector"
##   .. ..$ gdpPercap: list()
##   .. .. ..- attr(*, "class")= chr  "collector_double" "collector"
##   ..$ default: list()
##   .. ..- attr(*, "class")= chr  "collector_guess" "collector"
##   ..- attr(*, "class")= chr "col_spec"
# display few row in table
GapTxt %>%
  head() %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap
Congo, Dem. Rep. Africa 2002 44.966 55379852 241.1659
Congo, Dem. Rep. Africa 2007 46.462 64606759 277.5519
Lesotho Africa 1952 42.138 748747 298.8462
Guinea-Bissau Africa 1952 32.500 580653 299.8503
Congo, Dem. Rep. Africa 1997 42.587 47798986 312.1884
Eritrea Africa 1952 35.928 1438760 328.9406

Similar to .csv format, the structure of the data read from file is not the same as the data written to file, the factors continent and country are now character, although the table is still the same. We can also verify this from the plots.

P3 <- GapTxt %>%
 ggplot( aes(continent,gdpPercap)) +  geom_boxplot() +
  scale_y_log10() + ggtitle("Data read from file")

grid.arrange(GapOrdered,P3,ncol=2,top="Comparing the boxplots otained from data written to file and the one from data read from file")

In addition, we can write data into a file in such a way that each of the columns are separated by tab. This is the TSV format, meaning tab separated format

read_delim() and write_delim() separated by tab

Writing data to file:

write_delim(NewGapReorder,"NewGap.tsv",delim = "\t")  # writes the data to file

Reading data from file:

GapTab <- read_delim("NewGap.tsv", delim="\t")  # reads the gapminder file in csv format
## Parsed with column specification:
## cols(
##   country = col_character(),
##   continent = col_character(),
##   year = col_integer(),
##   lifeExp = col_double(),
##   pop = col_integer(),
##   gdpPercap = col_double()
## )
str(GapTab) # display structure of data
## Classes 'tbl_df', 'tbl' and 'data.frame':    1404 obs. of  6 variables:
##  $ country  : chr  "Congo, Dem. Rep." "Congo, Dem. Rep." "Lesotho" "Guinea-Bissau" ...
##  $ continent: chr  "Africa" "Africa" "Africa" "Africa" ...
##  $ year     : int  2002 2007 1952 1952 1997 1952 1952 1957 1952 1957 ...
##  $ lifeExp  : num  45 46.5 42.1 32.5 42.6 ...
##  $ pop      : int  55379852 64606759 748747 580653 47798986 1438760 20092996 813338 2445618 1542611 ...
##  $ gdpPercap: num  241 278 299 300 312 ...
##  - attr(*, "spec")=List of 2
##   ..$ cols   :List of 6
##   .. ..$ country  : list()
##   .. .. ..- attr(*, "class")= chr  "collector_character" "collector"
##   .. ..$ continent: list()
##   .. .. ..- attr(*, "class")= chr  "collector_character" "collector"
##   .. ..$ year     : list()
##   .. .. ..- attr(*, "class")= chr  "collector_integer" "collector"
##   .. ..$ lifeExp  : list()
##   .. .. ..- attr(*, "class")= chr  "collector_double" "collector"
##   .. ..$ pop      : list()
##   .. .. ..- attr(*, "class")= chr  "collector_integer" "collector"
##   .. ..$ gdpPercap: list()
##   .. .. ..- attr(*, "class")= chr  "collector_double" "collector"
##   ..$ default: list()
##   .. ..- attr(*, "class")= chr  "collector_guess" "collector"
##   ..- attr(*, "class")= chr "col_spec"
# display few rows of data in table
GapTab %>%
  head() %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap
Congo, Dem. Rep. Africa 2002 44.966 55379852 241.1659
Congo, Dem. Rep. Africa 2007 46.462 64606759 277.5519
Lesotho Africa 1952 42.138 748747 298.8462
Guinea-Bissau Africa 1952 32.500 580653 299.8503
Congo, Dem. Rep. Africa 1997 42.587 47798986 312.1884
Eritrea Africa 1952 35.928 1438760 328.9406

Here also, the structure of the data read from file is not the same as the data written to file, the factors continent and country are now character, although the table is still the same. Let display a boxplot of the data side-by-side.

P4 <- GapTab %>%
 ggplot( aes(continent,gdpPercap)) +  geom_boxplot() +
  scale_y_log10() + ggtitle("Data read from file")

grid.arrange(GapOrdered,P4,ncol=2,top="Comparing the boxplots otained from data written to file and the one from data read from file")

readRDS and saveRDS()

Writing to file:

saveRDS(NewGapReorder,"NewGap.rds")  # writes the data set in compressed format (.rds format)

This functions writes data to file in a compressed format.

GapRDS <- readRDS("NewGap.rds")  # reads from file

str(GapRDS) # display the structure
## Classes 'tbl_df', 'tbl' and 'data.frame':    1404 obs. of  6 variables:
##  $ country  : Factor w/ 142 levels "Afghanistan",..: 28 28 74 53 28 42 88 74 18 42 ...
##  $ continent: Factor w/ 5 levels "Africa","Oceania",..: 1 1 1 1 1 1 4 1 1 1 ...
##  $ year     : int  2002 2007 1952 1952 1997 1952 1952 1957 1952 1957 ...
##  $ lifeExp  : num  45 46.5 42.1 32.5 42.6 ...
##  $ pop      : int  55379852 64606759 748747 580653 47798986 1438760 20092996 813338 2445618 1542611 ...
##  $ gdpPercap: num  241 278 299 300 312 ...
# display few rows of data in table
GapRDS %>%
  head() %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap
Congo, Dem. Rep. Africa 2002 44.966 55379852 241.1659
Congo, Dem. Rep. Africa 2007 46.462 64606759 277.5519
Lesotho Africa 1952 42.138 748747 298.8462
Guinea-Bissau Africa 1952 32.500 580653 299.8503
Congo, Dem. Rep. Africa 1997 42.587 47798986 312.1884
Eritrea Africa 1952 35.928 1438760 328.9406

Unlike, the previous format we have seen, this format preserves the structure of the data when reading from file. Let us confirm that the level are still in the same order as it was written to the file.

levels(GapRDS$continent) # displays the levels of the selected factor in proper order
## [1] "Africa"   "Oceania"  "Europe"   "Asia"     "Americas"

Observe that the ordering is also preserved. Yay! Let us check the boxplots:

P5 <- GapRDS %>%
 ggplot( aes(continent,gdpPercap)) +  geom_boxplot() +
  scale_y_log10() + ggtitle("Data read from file")


grid.arrange(GapOrdered,P5,ncol=2,top="Comparing the boxplots otained from data written to file and the one from data read from file")

The ordering is also preserved in the figures.

dput() and dget() functions

Writing to file:

dput(NewGapReorder,"MyGapMinderDput.txt")  # writes data to file

Reading from file:

GapDput <- dget("MyGapMinderDput.txt")  # reads data from file

str(GapDput) # displays structure
## Classes 'tbl_df', 'tbl' and 'data.frame':    1404 obs. of  6 variables:
##  $ country  : Factor w/ 142 levels "Afghanistan",..: 28 28 74 53 28 42 88 74 18 42 ...
##  $ continent: Factor w/ 5 levels "Africa","Oceania",..: 1 1 1 1 1 1 4 1 1 1 ...
##  $ year     : int  2002 2007 1952 1952 1997 1952 1952 1957 1952 1957 ...
##  $ lifeExp  : num  45 46.5 42.1 32.5 42.6 ...
##  $ pop      : int  55379852 64606759 748747 580653 47798986 1438760 20092996 813338 2445618 1542611 ...
##  $ gdpPercap: num  241 278 299 300 312 ...
# display few rows in table
GapDput %>%
  head() %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap
Congo, Dem. Rep. Africa 2002 44.966 55379852 241.1659
Congo, Dem. Rep. Africa 2007 46.462 64606759 277.5519
Lesotho Africa 1952 42.138 748747 298.8462
Guinea-Bissau Africa 1952 32.500 580653 299.8503
Congo, Dem. Rep. Africa 1997 42.587 47798986 312.1884
Eritrea Africa 1952 35.928 1438760 328.9406

The structure of the data is also preserved with this format. How about we plot a boxplot of the data read from file and compare with the boxplot of the data written to file?

P6 <- GapDput %>%
 ggplot( aes(continent,gdpPercap)) +  geom_boxplot() +
  scale_y_log10() + ggtitle("Data read from file")

grid.arrange(GapOrdered,P6,ncol=2,top="Comparing the boxplots otained from data written to file and the one from data read from file")

These figures are identical, which confirms that the structure of the data is preserved.

Part 3: Visualization design

We begin by loading the necessary libraries required for this exercise.

suppressPackageStartupMessages(library(scales))
suppressPackageStartupMessages(library(plotly))
FirstAttempt <- gapminder %>%
            ggplot(aes(pop,lifeExp)) + geom_point(aes(color=gdpPercap), alpha = 0.3) 

Attempt1 <- FirstAttempt  +  scale_x_log10() + ggtitle("Attempt 1")
Attempt1 

Attempt2 <-  FirstAttempt +  # loads the previous plot
            scale_color_continuous(trans="log10",       # makes the colorign continuous
                                   breaks = 5^(1:10),             # labels the legend in multiples of 5
                                   labels = comma_format()        # wtites the legend label in comma separated format
                                    ) + 
            scale_y_continuous(breaks = 10*(1:10)) + # labels the y-axis in multiples of 10
            scale_x_continuous(trans = "log10",   # puts the x-axis in log10 scale
                               labels = comma_format(),  # put the x-label in comma format
                               ) + ggtitle("Attempt 2: scale_color_continuous")
Attempt2  

Attempt3 <-  FirstAttempt +  # loads the previous plot
            scale_color_distiller(trans="log10",       # makes the colorign continuous
                                   breaks = 5^(1:10),             # labels the legend in multiples of 5
                                   labels = comma_format(),        # wtites the legend label in comma separated format
                                    palette = "Reds") + 
            scale_y_continuous(breaks = 10*(1:10)) + # labels the y-axis in multiples of 10
            scale_x_continuous(trans = "log10",   # puts the x-axis in log10 scale
                               labels = comma_format(),  # put the x-label in comma format
                               )  + ggtitle("Attempt 3: scale_color_distiller")
                  
Attempt3 

Attempt4 <-  FirstAttempt +  # loads the previous plot
            scale_color_viridis_c(trans="log10",       # makes the colorign continuous
                                   breaks = 5^(1:10),             # labels the legend in multiples of 5
                                   labels = comma_format()        # wtites the legend label in comma separated format
                                   ) + 
            scale_y_continuous(breaks = 10*(1:10)) + # labels the y-axis in multiples of 10
            scale_x_continuous(trans = "log10",   # puts the x-axis in log10 scale
                               labels = comma_format(),  # put the x-label in comma format
                               )   + ggtitle("Attempt 4: scale_color_viridis_c")
                  
Attempt4

Putting all the attmepts together, we have

 grid.arrange(Attempt1,Attempt2,Attempt3,Attempt4,ncol=2,top="Different Attempts")

Now let us convert each of our attempts to a plotly

plotly1 <- ggplotly(Attempt1)
## Warning: replacing previous import by 'Rcpp::evalCpp' when loading 'later'
## Warning: replacing previous import by 'shiny::validateCssUnit' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::br' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::tags' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::div' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::h1' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::h2' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::h3' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::h4' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::h5' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::h6' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::knit_print.html' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::tagSetChildren' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::includeScript' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::em' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::tagAppendChild' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::is.singleton' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::includeHTML' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::includeMarkdown' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::code' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::tagList' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::a' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::tagAppendAttributes' when
## loading 'crosstalk'
## Warning: replacing previous import by 'shiny::singleton' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::hr' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::p' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::suppressDependencies' when
## loading 'crosstalk'
## Warning: replacing previous import by 'shiny::tagAppendChildren' when
## loading 'crosstalk'
## Warning: replacing previous import by 'shiny::includeText' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::pre' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::span' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::withTags' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::htmlTemplate' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::img' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::tag' when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::includeCSS' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::knit_print.shiny.tag' when
## loading 'crosstalk'
## Warning: replacing previous import by 'shiny::knit_print.shiny.tag.list'
## when loading 'crosstalk'
## Warning: replacing previous import by 'shiny::strong' when loading
## 'crosstalk'
## Warning: replacing previous import by 'shiny::HTML' when loading
## 'crosstalk'
plotly1
plotly4 <- ggplotly(Attempt4)
plotly4

We observe from these figures that the plotly() allows for more interaction with the figures, for example, we can zoom in and out, select part of the figure, save figure as .png file, plus many other interactiveness. It evens shows the value of the variables in the plot when the mouse is placed over each point in the graph.

Let do some activities with the theme() function. First, we plot some figures form the gapminder data, after which we make it more presentable.

lifeExp_vs_Gdp <- gapminder %>%
    ggplot(aes(gdpPercap, lifeExp)) + geom_point(aes(color=continent)) + 
    scale_x_log10() + 
    facet_wrap(~continent) +
    ggtitle( "Life Expectancy vs Gdp per capital for different continents")

lifeExp_vs_Gdp

lifeExp_vs_Gdp +
    theme_bw() +
    theme(axis.text = element_text(size=16),
          strip.text = element_text(size=16,color = "blue"),
          strip.background = element_rect(fill = "orange"),
          panel.background = element_rect(fill = "pink"))

Let us plot directly using the ‘plot_ly()’ function

gapminder %>%
  plot_ly(x = ~gdpPercap, 
        y = ~pop, 
        type = "scatter",
        colors = 'red',
        mode = "markers",
        marker=list(color="green" , size=5 , opacity=1.5) ,
        opacity = 0.5) %>% 
    layout(xaxis = list(type = "log"),yaxis = list(type = "log"))
gapminder %>%
  plot_ly(x = ~lifeExp, 
        y = ~gdpPercap, 
        type = "scatter",
        colors = 'red',
        mode = "markers",
        marker=list(color="red" , size=10 , opacity=0.5) ,
        opacity = 0.5) %>% 
    layout(xaxis = list(type = "log"),yaxis = list(type = "log"))

Part 4: Writing figures to file

NewGap_Boxplot <-NewGap %>%
  mutate(continent= fct_reorder(continent,gdpPercap,max)) %>%
  arrange(gdpPercap) %>%
  ggplot( aes(continent,gdpPercap)) +   geom_boxplot(fill='pink') +
  scale_y_log10()  + ggtitle("Boxplot of population for each continent ordered by the median population")


ggsave("Boxplot.png", NewGap_Boxplot)
## Saving 7 x 5 in image
ggsave("Boxplot2.png", plot = NewGap_Boxplot, width = 2, height = 2)
ggsave("Boxplot3.png", plot = NewGap_Boxplot, width = 5, height = 2, limitsize =TRUE, dpi = 300) 
ggsave("Boxplot4.pdf", plot = NewGap_Boxplot, width = 5, height = 2, limitsize =TRUE, dpi = 300)

graph

But I want to do more!

Let us extract the rows for some countries in the gapminder data:

Sport_Gap <- gapminder %>%
    filter(year == "1952") %>%
    filter(country %in% c("Nigeria", "Canada", "India", "Sweden", "Brazil")) %>%
    mutate(Sports = factor(c("Soccer","Ice hockey","Cricket","Biking","Baseball"))) %>%
    droplevels()
Sport_Gap %>%
  knitr::kable()
country continent year lifeExp pop gdpPercap Sports
Brazil Americas 1952 50.917 56602560 2108.9444 Soccer
Canada Americas 1952 68.750 14785584 11367.1611 Ice hockey
India Asia 1952 37.373 372000000 546.5657 Cricket
Nigeria Africa 1952 36.324 33119096 1077.2819 Biking
Sweden Europe 1952 71.860 7124673 8527.8447 Baseball 
Sport_Gap %>%
  str
## Classes 'tbl_df', 'tbl' and 'data.frame':    5 obs. of  7 variables:
##  $ country  : Factor w/ 5 levels "Brazil","Canada",..: 1 2 3 4 5
##  $ continent: Factor w/ 4 levels "Africa","Americas",..: 2 2 3 1 4
##  $ year     : int  1952 1952 1952 1952 1952
##  $ lifeExp  : num  50.9 68.8 37.4 36.3 71.9
##  $ pop      : int  56602560 14785584 372000000 33119096 7124673
##  $ gdpPercap: num  2109 11367 547 1077 8528
##  $ Sports   : Factor w/ 5 levels "Baseball","Biking",..: 5 4 3 2 1